Transcript PREVENTION

BACK PAIN AND LUMBAR STENOSIS
IN OLDER ADULTS
RESEARCH GROUP
University of Pittsburgh
Shervadalonna Brown, MD
Jane A. Cauley, DrPH
William F. Donaldson, MD
James D. Kang, MD
Douglas Musgrave, MD
Terence Starz, MD
Mark Chirumbole, BS
Anthony DeLitto, PhD
Julie Fritz, PhD
Lewis H. Kuller, MD
Molly T. Vogt, PhD
University of California at San Francisco
Michael Nevitt, PhD
Ria San Valentin, MD
Lisa Palermo, MS
Georgetown University
William C. Lauerman, MD
Dartmouth Medical Center
Brett Hanscom, MS
James Weinstein, DO
Washington University, St Louis
David Rubin, MD
INTRODUCTION
80% of US population experience one or
more episodes of low back pain during
lifetime.
One of leading causes for physician office
visits and for filing disability claims.
During last 3 decades disability claim rate
has increased 13 fold.
Annual medical costs related to back pain
are estimated to be $8 - $18 billion.
INTRODUCTION
Low back pain (LBP) affects 90% of individuals.
Initial episode of
LBP
Persistent LBP after
4-6 weeks-10%
Recovery in 4-6
weeks 90%
Recurrence in
one year- 30%
LBP impacts quality of life & health care
expenditures.
INTRODUCTION
National guidelines have recommended
the use of analgesics as the primary
pharmacologic treatment for LBP.
The choice of analgesic agent has major
implications for health care costs.
UPMC Health Plan (Commercial)
• 17,228 (14.8% of total) health plan
members had at least one claim for
service (pharmacy, inpatient, outpatient,
laboratory, and physical /occupational
therapy) for LBP management.
Total cost = $6,419,696
• 9,566 (56% of members with LBP claims)
had pharmacy claims for narcotics,
NSAID’s, Cox2’s or other analgesics.
Total cost = $1,403,837
UPMC-HP Member Resource
Utilization for LBP
40%
35%
Narcotics
X-rays
NSAIDs
30%
25%
20%
15%
10%
5%
0%
MRI PT/OT
Cox2s
analgesics
Narcotic costs for UPMC-HP
members with LBP or cancer
$
1000000
800000
600000
400000
200000
0
LBP
48% of total narcotic costs
attributed to members with
LBP, 21% to members with
cancer
Cancer
Utilization Pattern of Pain Medications
among LBP patients in UPMC-HP
Narcotics+ other
analgesics
Narcotics
alone
Narcotics+nonselective NSAID
analgesic alone
Cox2 alone
NSAIDs alone
INTRODUCTION
Back pain in adult patients linked with:
• lifestyle factors (smoking, obesity,
physical activity, education)
• anatomic abnormalities of lumbar spine
Back pain in the elderly related to:
• degenerative changes due to aging
• lifestyle less important
Back pain in older persons
Increasing age is associated with an increase in
musculoskeletal symptoms
In the US back pain is the 3rd most frequent
symptom reported to MDs by persons 75+ years
17% of back problem visits occur in those
aged 65+ years
BUT
neither prevalence nor health burden
is known
Prevalence of back pain in older
persons
# studies
% prevalence
Community
9
13 - 49
Primary practice
3
23 - 51
Nursing home
1
40
Bressler, et al. Spine 1999
Prevalence of back pain in older
persons
Prevalence seems to decrease a little with age
Women usually report a higher prevalence than
men
A major problem is the definition of back pain
“no gold standard”
No studies of the validity/reliability of dx orthopaedic
testing procedures, no validity studies of clinical or
self report of location of back pain
Bressler, et al. Spine 1999
Patient factors contributing to the
variability of prevalence of back pain in
older persons
• cognitive impairment
• depression
• decreased pain perception
• increased pain tolerance
• comorbid conditions
• decreased physical activity
• resignation to aging effects
• selective participation in studies
Overall seems likely that back pain is often
under-reported
Relationship between history of CVD at baseline
and back problems at the 3rd clinic visit
Age-adj OR (95% CI)
______________________________________________
Back pn since 1st clin vis
none
mild/mod
severe
1.0
1.3 (1.0, 1.6)
2.6 (1.7, 4.0)
One + days of lim act
due to back pain
2.3 (1.6, 2.3)
One + days in bed
due to back pain
1.2 (0.6, 2.3)
Vogt, et al, Spine 1997
Odds ratio for back pain at baseline in SOF
women (65+ yrs) by estrogen usage
1.6
1.4
*
*
*
*
1.2
Never
Former
Current
1
0.8
0.6
0.4
0.2
0
baseline
follow-up
Visit
Musgrave, et al. Spine 2001
Causes of back pain in older patients
Acute (< four weeks)
lumbar strain/sprain
osteoporotic fracture, vertebral or pelvic
abdominal aortic aneurysm
Subacute/Chronic (> four weeks)
degenerative disc and joint disease
malignancy
fibromyalgia
polymyalgia rheumatica
Parkinson’s disease
Predictors of chronicity of low back
pain in adults (n=1246)
Better function at Chronic LBP
12 weeks n=1150
n=96
Age (yrs)
42.5
44.6*
Nonwhite race (%)
14
23*
Income>$20K/yr
73
50*
10.5
17.4*
22
47*
Baseline Roland
score
Baseline sciatica (%)
Carey, et al, Spine 2000
Primary location of pain
1. Lower back pain alone
2. Pain radiating into buttocks and leg
* upper anterior thigh/groin
* lateral hip
* below knee
Malignant, infectious or visceral pain is constant
whatever position of body. Mechanical, myofascial
or degenerative pain varies by body position usually lessens when person
is supine
Radicular pain
spinal nerve entrapment by disc herniation
or spinal stenosis
pain in leg, paresthesia, weakness
Causes of leg pain in older patients
True radicular pain
lumbar stenosis
lumbar disc herniation
Pseudosciatica
trochanteric bursitis
osteoarthritis of the hip
diabetic neuropathy
ANATOMY
Normal human spine is lordotic in the
lumbar region. During typical movements
upper lumbar vertebrae - posterior shear
lower lumbar vertebrae - anterior shear
Stability maintained by facet joints,
intervertebral discs, ligaments, related
muscle groups
PATHOLOGY
Degenerative changes in the lumbar spine
disc degeneration
vertebral compression deformities
ligamentous laxity
deterioration of facet joint cartilage
May cause instability and subluxation of one
vertebra on another (degenerative
spondylolisthesis)
Anterolisthesis at L4-L5
PATHOLOGY
Backward slippage (retrolisthesis) is
generally believed to be asymptomatic and
of little clinical significance.
Forward slippage (anterolisthesis) may
result in narrowing of vertebral canal and
neural foramina (spinal stenosis) leading to
development of chronic back pain (with or
without leg pain). Compression of L5 spinal
nerve may be involved.
PATHOLOGY
When LS joint is particularly stable, L4
and L5 are more vulnerable to stress
forces. If degenerative changes have
occurred, anterolisthesis at L4 is more
likely.
Clinical symptoms associated with
anterior subluxation
at L4-L5
at L3-L4
80%
10-20%
PATHOLOGY
Spinal stenosis symptoms:
back pain progressing to leg pain
functional independence deteriorates
reduced ability to walk
reduced ability to carry out ADLs
Symptoms often episodic, no natural
resolution over time
EPIDEMIOLOGY
Several clinical and cadaveric studies
suggest that anterolisthesis is
5 times more common in women vs men
2-4 times more common in blacks than whites
4 times more prevalent in diabetics
3 times more common in oophorectomized
women compared to controls
Prevalence of lumbar listhesis
(L3-S1) in elderly white women (SOF)
% prevalence
listhesis defined as subluxation > 3mm
45
40
35
30
25
20
15
10
5
0
anterolisthesis
retrolisthesis
p for trend = 0.027
p for trend = 0.75
65-69
70-74
75-79
Age in years
80+
CLINICAL RELATIONSHIPS
Relationship between radiographic
abnormalities and spinal symptoms is
unclear.
People with no back pain show disc
abnormalities (64%), stenosis (7%) and
anterolisthesis (7%) (Boden, JBJS 1990,
Jensen NEJM 1994 ).
Not known whether people with sub-clinical
disease later develop symptoms.
Veteran’s Health Study
n= 428 men
% of cohort
45
40
35
30
25
20
15
10
5
0
LBP only
LBP+LP to thigh
LBP+LP below knee
Selim, et al. Spine 1998
Veteran’s Health Study
Medic use
MRI
Surgery
LBP alone
1.0
1.0
1.0
LP to thigh
1.5
3.2
0.9
(0.7,3.1)
(1.5,6.7)
(0.3,3.0)
1.8
3.5
3.7
(1.0,3.4)
(1.9,6.5)
(1.7,8.1)
5.1
6.8
3.9
(1.2,22.9)
(2.7,17.2)
(1.3,11.4)
LP below knee
(-ve SLR)
LP below knee
(+ve SLR)
Selim, et al. Spine 1998
SF-36 scores for men with LBP enrolled in the
Veteran’s Health Study
70
p for trend <0.05 for all domains
60
Score
50
40
30
LBP only
LBP/LP to thigh
LBP/LP below knee (-ve SLR)
LBP/LP below knee (+ve SLR)
20
10
0
PF
RP
BP
GH
VT
MH
SF
RE
Selim, et al. Spine 1998
Distribution of lower back and leg pain symptoms w/in
last month among white WHI women aged 50 years and older
60
% of cohort
50
40
30
20
10
0
No LBP
n=295
LBP only
n=47
LBP+LP
n=182
LBP+LP impr
by sitting
n=49
Vogt et al. J Gerontol 2002
SF-36 scores for white women enrolled in WHI
(adjusted for age and BMI)
Vogt et al. J Gerontol 2002
100
90
Score
80
70
60
50
no LBP
LBP
LBP/LP
LBP/LP improved by sitting
40
30
PF
RP
BP
GH
VT
MH
SF
RE
Relationship of race to prevalence and
use of health care resources for LBP
Random digit dialing + structured interview
4,437 households in NC
8067 individuals
Whites (%)
AAmer (%)
Prev acute LBP last yr
8.3 (7.3, 9.3)
5.2 (3.8, 6.6)
Prev chronic LBP last yr
4.1 (3.4, 4.7)
3.0 (2.0, 4.0)
36
59
Prev seeking care
Carey, et al, Spine 1996
Relationship of race to prevalence and
use of health care resources for LBP
Cohort study, random group of health care providers
Whites
AAmer
5.25
5.92
<0.01
Disability score
11
12.1
0.01
X-rays (%)
49
40
0.05
Other imaging
10
6
0.05
Pain score
p
Carey, et al, 2000
Elderly African American women (SOF)
reporting back pain during previous four
weeks
N=470
severe LBP
7%
moderate LBP 20%
50%
no LBP
mild LBP 23%
% frequency
Back/leg symptoms in women aged 65 years and older
during month prior to clinic visit (white women enrolled
in WHISTEN, black women enrolled in SLIP)
100
90
80
70
60
50
40
30
20
10
0
36.1
28.7
9.8
21.6
54.1
49.7
White women
N=399
Black women
N=470
Back pain, with leg
symptoms
Back pain, no leg
symptoms
No back pain
Prevalence of lumbar listhesis (L3-S1)
in black elderly women by age
listhesis defined as subluxation > 3mm
% prevalence
80
p for trend = 0.095
70
60
50
40
Anterolisthesis
Retrolisthesis
30
20
10
p for trend = 0.207
0
65-69
70-74
75-79
Age in years
80+
% prevalence of listhesis among
women 65 years and older
Antero
Retro
White
Black
White
Black
L3-L4
4
13
6
1
L4-L5
20
36
4
2
L5-S1
9
30
7
3
L3-S1
29
58
14
4
Vogt, et al, The Spine J 2002
Effect of back pain & leg pain on daily life
of black women during previous month
6
expressed as age-adj odds ratio using back
pain only as the reference - all p<0.001
5
4
3
2
1
0
mood
walk/move
sleep
work
recreation
enjoy
Vogt, et al, The Spine J 2002
PREVENTION
Because most people experience LBP
during their lifetime, the distinction between
primary and secondary prevention is blurred.
• which interventions can prevent
occurrence of LBP?
• which interventions can prevent
development of chronic LBP?
PREVENTION
Evidence-based medicine categories
Level A - strong consistent - multiple RCTs
Level B - moderate - one RCT + multiple CCTs
Level C - limited - one CCT
Level D - no evidence
PREVENTION
Lumbar supports
Level A - ve
•provide support
• remind to lift properly
• intra-abdom pressure and
• intradiscal pressure
RCTs negative
CCTs positive – reduce incidence
of LBP and back injury
PREVENTION
Back Schools and Education Level A -ve
• provide knowledge about body
mechanics, stress, exercise
• aim to influence behavior
9 RCTs - most are negative
5 CCTs - positive
PREVENTION
Exercises
Level A + ve
• strengthen back muscles
• increase blood supply
• improve mood and alter perception
of pain
6 RCTs – reduced pain and sick
leave
PREVENTION
Ergonomics Level D - ve
• job related interventions
No RCTs or CCTs
PREVENTION
Risk Factor Modification Level D - ve
• individual (weight, strength, smoking)
• biomechanical (lifting, posture)
• psychosocial (job control, job
dissatisfaction, depression)
No RCTs or CCTs
Review of 47 epidemiologic studies
concluded that smoking may be a ‘weak risk
indicator and not a cause of low back pain’
Le-Bouef-Yde Spine 1999
Smoking may have a systemic effect on the
musculoskeletal system - associated with
generalized pain.
Biological basis unknown - neuroendocrine
effect?
Decrement in SF-36 scores (compared to
age-sex specific norms) for patients with
spinal problems by smoking status
PF
RP
BP
HP
MH
EF
SF
RE
SF-36 score
0
-10
-20
-30
-40
-50
Smokers (n = 4249)
-60
Non-smokers (n = 21206)
-70
General population in US
-80
Vogt, et al, Spine 2002
PREVENTION
Currently only exercise seems to be
helpful in prevention of LBP.
Consistent evidence – Level A.
Linton, van Tulder, Spine 2001
PREVENTION
Why the disappointing results?
• small studies, low power, short follow-up,
variation in intervention, varying outcome
• natural course of back pain, hard to define
and categorize, multi-factorial causation
• single modal programs studied mostly,
maybe multi-dimensional approach needed
• timing, compliance